Pub 2010

Article, Accepted Version
Söhngen, Bernhard; Kayser, Jan

Design of bank and bottom protection
On Course - PIANC magazine
Verfügbar unter/Available at: https://hdl.handle.net/20.500.11970/100652
Vorgeschlagene Zitierweise/Suggested citation:

Söhngen, Bernhard; Kayser, Jan (2010): Design of bank and bottom protection. In: On
Course - PIANC magazine 140. S. 5-16.
Standardnutzungsbedingungen/Terms of Use:
Die Dokumente in HENRY stehen unter der Creative Commons Lizenz CC BY 4.0, sofern keine abweichenden
Nutzungsbedingungen getroffen wurden. Damit ist sowohl die kommerzielle Nutzung als auch das Teilen, die
Weiterbearbeitung und Speicherung erlaubt. Das Verwenden und das Bearbeiten stehen unter der Bedingung der
Namensnennung. Im Einzelfall kann eine restriktivere Lizenz gelten; dann gelten abweichend von den obigen
Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte.
Documents in HENRY are made available under the Creative Commons License CC BY 4.0, if no other license is
applicable. Under CC BY 4.0 commercial use and sharing, remixing, transforming, and building upon the material
of the work is permitted. In some cases a different, more restrictive license may apply; if applicable the terms of
the restrictive license will be binding.
Powered by TCPDF (www.tcpdf.org)

Autorenfassung
Söhngen, Kayser: Design of bank and bottom protection. New design principles for the necessary riprap stone
sizes and revetment thicknesses for protection against wave and current attack caused by inland vessels,
2010
Erstveröffentlichung in PIANC Magazine no° 140 (2010), S. 5-16.

Für eine korrekte Zitierbarkeit ist die Seitennummerierung
der Originalveröffentlichung für jede Seite kenntlich gemacht. S. 5
Design of bank and bottom protection

New design principles for the necessary riprap stone sizes and revetment
thicknesses for protection against wave and current attack caused by in-
land vessels
By Bernhard Söhngena, Jan Kayserb

Federal Waterways Engineering and Research Institute, Kussmaulstr. 17, 76187 Karlsruhe, Germany
a bernhard.soehngen@baw.de
b jan.kayser@baw.de
Key words: Bank protection, hydraulic design, geotechnical design, riprap, revetment
Mots-clefs: Defenses de berge, dimensionnement hydraulique, dimensionnement geotechnique, enro-

chement, revetement
1 Hydraulic design- determining of riprap stone size
1.1 Changed boundary conditions and investigations regarding the new 'GBB'
The German as well as the western European inland vessel fleet has been strongly changed during
the last decades. One reason for this change is the stronger competition of inland water transport to
rail and road since EU law has liberalised freight rates in 1993. Another reason is the demand for
higher safety requirements for modern vessels and units, together with ongoing developments in
the construction of new inland vessels. Therefore, vessels are nowadays on average much more
powered than about 10 years ago. Moreover, modern inland vessels' loading capacity is continuous-
ly increasing. For these reasons, modern inland vessels are able to reach the so called critical ship
speed vcrit, especially in canals, which in former times was normally not the case because of lower
engine power and smaller blockage ratios of the smaller ships.
For example, in field tests performed by the German Federal Waterways Engineering and Research
Institute (BAW) on the Wesel-Datteln-Canal (WDC), which is developed as a standard trapezoidal
profile, the 'Main' motor ship (MS), powered by an approximately 900kW engine, reached the criti-
cal speed range not only when sailing close to the bank - which is already being observed also for
conventional 'Europe Ships' - but also when moving in the centre of the canal. More powered inland
cargo vessels (GMS), as the modern MS 'Aviso II' with an engine power of about 1300 kW, can sail -
theoretically - even above the critical speed range in canals. However, the supercritical speed range
is generally not relevant for the design of bed and bank protections, because drawdown, wave
-1-
Autorenfassung
2010
heights, squat and necessary power increase strongly when reaching vcrit and the ship cannot sail
much faster than with vcrit. In addition, critical sailing conditions can be recognised without any
doubt by the wave formation,
Söhngen, Kayser: Design of bank and bottom protection. S. 6

PIANC Magazine 140, S. 5-16.
especially by a breaking stern wave. This is why skippers avoid the supercritical range - further-
more as there is the risk to run aground.
Nevertheless, a noticeable share of vessels moves in the critical ship speed range - as observed in
2002 at km 82 of the Dortmund-Ems-Canal (DEC); see Figure 1. Critical speed conditions may also
be achieved unintentionally, e.g. when the vessel approaches the bank or in case of cross-sectional
changes. Together with nowadays permitted larger draughts of 2.8m this had the effect that the
canal lining was exposed to much higher ship induced loads than expected when the German Code
of Practice 'Use of Standard Construction Methods for Bank and Bottom Protection on Waterways'
(MAR) ['Merkblatt zur Anwendung von Regelbauweisen für Böschungs- und Sohlensicherungen
von Wasserstraßen', MAR] was published in 1993 and when the German 'Guidelines for standard
cross-sections of navigation canals" [Richtlinien für die Regelquerschnitte von Schifffahrtskanälen']
were published in 1994.
Fig. 1: Frequency distribution of ship speed vs related to vcrit of all laden vessels moving up-
stream and downstream on the Dortmund-Ems-Canal around km 82 - as observed in
2002
The obviously increased hydraulic impacts on the revetments that were designed according to the
a. m. MAR provoked stone displacements larger than expected - and this only 3 to 5 years after the
extension of the above-mentioned Dortmund-Ems-Canal stretch had been completed. Stone dis-
placements mainly occur in the zone of fluctuating water Ievel. Now and in the foreseeable future,
-2-
Autorenfassung
2010
they do not yet endanger the bank stability. However, they should be limited in course of mainte-
nance in order to guarantee future stability.
In order to be able to react to revetment Load changes, a working group was established in 1992
upon the initiative of the BAW; this working group comprised representatives of the WSV (German
Federal Waterways and Shipping Administration) and of the BAW with the intention of compiling
bases for designing bank and bottom revetments, starting from the existing knowledge. This work
resulted in the 'Principles for the Design of Bank and Bottom Protection for Inland Waterways'
(GBB) ('Grundlagen zur Bemessung von Böschungs- und Sohlensicherungen an Binnenwasser-
straßen'], published - in English - in BAW Mitteilungsblatt (newsletter) No. 88 in 2004 and in a re-
vised version of MAR available since end of 2008.
A first version of the new design principles was produced in 1999. It was based mostly on the work
of Fuehrer and Römisch (1985) from the former 'Forschungsanstalt für Schifffahrt, Wasser- und
Grundbau' (FAS) ('Research Institute for Navigation, Hydraulic and Foundation Engineering'], Ber-
lin, regarding the design against ship-induced waves and return flow, the results of PIANC Working
Group 21, published in 1992 in the Report on the 'Design of Flexible Revetments', the publication of
Przedwojski et al. (1995) on 'River Training Techniques' and concerning the hydraulic loads from
propeller wash, on the numerous works publications of Fuehrer and Römisch (1975, 1985, 1989
and 1994 ). First applications of these design principles and comparative calculations the equations
given in the above-mentioned FAS and PIANC publications showed significant differences in the
relevant wave heights for the crosssection ratios according to the German standard canal profiles
for loads provoked by modern in land vessels - i.e. for ratios of the cross-sectional area of a canal to
the submerged cross-sectional area of a vessel (n) of about 5-7. Especially it seemed that the design
principles - resulting amongst others from model tests with different vessel types including sea-
going vessels, sea-going vessels suitable for inland navigation and passenger ships, mostly executed
with high n-ratios - given by FAS

for sailing within the critical ship speed range in canals with standard profiles, were set too low as
they were not suitable for explaining the observed revetment damages. Furthermore, differences
became obvious concerning the definition of the critical ship speed. According to the a.m. PIANC
report, which is based on Schijf's 1 D approach (1953), vcrit is derived from plotting the calculated
width-averaged drawdown h against the ship speed vs. This curve is rising and dropping down
again with increasing Δh. By this definition, vcrit is the maximum possible ship speed. In the FAS
publications, vcrit was defined by the arising of a breaking stern wave that can occur before reaching
vcrit according to Schijf's definition, especially in case of stern-loaded vessels.
-3-
Autorenfassung
2010
This gave rise to the above-mentioned site investigations performed at the Dortmund-Ems-Canal in
order to determine the speed at which inland vessels actually move nowadays in standard profile
canals; and additionally to new field tests with rented modern inland vessels (MS 'Main', length:
L = 105m, beam width: B = 11 .0m, draught: t8 = 1.7- 2.7m, installed power: P0 = 900kW, 'Mignon'
tug boat, L = 21m, B = 5m, t8 = 1.7m, P 0 = 300 kW) at design-relevant ship speeds close to vcrit and
sailing positions up to a ship path close to the banks. The purpose was to achieve design values
close to reality especially for ship-induced waves. The results of these investigations have shown
• that about half of the vessels observed moved at a higher speed than permitted, which led to the
recommendation to use vcrit as the scaling variable for the design ship speed instead of the per-
mitted speed
• that according to these measurements, some 2% of all vessels moved faster than 0.97 vcrit (see
Fig. 1), corresponding to about 1 or 2 vessels per day considering the current traffic on the
Dortmund-Ems-Canal. Therefore the usual design ship speed was increased from 0.9 vcrit as
proposed amongst others in the PIANC publications to 0.97 vcrit – particularly as the last-named
value was consistent with the MAR
• that, at critical ship speed and when sailing close to the bank, ship-induced waves can reach
1.0m with fully laden large inland cargo vessels (GMS), 1.1 m with empty/ballasted large inland
cargo vessels (GMS) and up to 1.3 m with tugs and
• that the slope supply flow- which is a flow next to the bank running with the vessel and occur-
ring especially in the case of breaking stern waves, forming a characteristic "tongue" in the
stern area as shown in Fig. 2 - was the cause for the largest stone displacements observed.
Fig. 2: Travel of MS 'Main' in the Wesel-Datteln-Canal in 2002 with the relevant hydraulic
Load forces shown on the water surface
Therefore, the existing approaches needed to be revised and finally resulted in the above-
mentioned GBB in 2004. Due to remaining uncertainties - especially resulting from the loads of rec-
reational traffic and exceptional vessels, e.g. pusher craft sailing alone, from bank revetments with
-4-
Autorenfassung
2010
very gentle slope gradients as occurring in natural rivers, and from loads of modern vessel propul-
sion and steering devices, e.g. bow thrusters or pump jet thrusters (to give some examples), for
which only few experiences and design approaches exist - it was not possible to recommend nei-
ther generally applicable design rules nor a code of practice on the basis of knowledge in 2004. For
this reason, GBB offers in some cases more than one design formula for the same impact. These do
not keep free planning engineers from taking up the responsibility to choose the 'right' design ap-
proaches in each single case and bring their own experience and appropriate literature if necessary.
Therefore, the GBB is not a classical body of design but it offers formulae for certain Load and re-
vetment types for which experience already exists.

In order to close existing gaps of knowledge, a test phase was started in 2004, applying GBB formu-
lae to several existing information of ship-induced loads, including literature data and several site
investigations in German canals and rivers, made in the past in the context of specific design prob-
lems and ongoing newer studies in the framework of a BAW R&D project to determine the state of
existing revetments together with the corresponding navigational loads. In this context a data basis
was formed, consisting of thousands of experiments. Additionally, new scale model tests were
commissioned to the 'Development Centre of Ship Design and Transport Systems' (DST) in Duis-
burg, accompanied and interpreted by the BAW, regarding canals that are significantly narrower
and wider than the standard cross-sections with bank slopes flatter and steeper than 1:3 and taking
into account not only usual inland going vessels but also single sailing pushers and recreational
boats. These investigations are still in progress and will probably be finished at the end of 2009.
Without dealing details one can state from the first results that the design rules in GBB are general-
ly on the safe side, but it is obvious that the GBB formulae should not be applied without re-
strictions to design cases for which so far only few experience exists.
Nevertheless, regarding the various experience for ship-induced loads in German standard canal
profiles, these results gave rise to the revision of the existing MAR on the basis of calculations for
typical impacts for these relations. It is available since the end of 2008 and was established by the
German Government in 2009. Additionally, the design principles of GBB, including all the experi-
ence gathered since 2004, were put into a user-friendly software tool, named 'GBBSoft', which is
available for engineers, researchers and training purposes on the BAW website (www.baw.de) at
the beginning of 2009, together with a new version of the GBB, which contains all the modified ap-
proaches, used in GBBSoft. An appropriate training is recommended and will be offered by the BAW
staff. Furthermore it is planned to adapt GBBSoft probably every year according to the increasing
state of knowledge. As a first step it is foreseen to modify the formulae concerning the slope supply
flow velocity, using the results of the a. m. model test, which show that significant velocities only
occur in case of breaking stern waves, so that a breaking criterion has to be included into the design
formulae.
-5-
Autorenfassung
2010
1.2 The new design principles
As the derivation or even the mere listing of the formulae given in GBB would by far exceed the
extent of this publication (interested parties are invited to consult the literature given in the GBB),
only the core aspects of the new GBB are briefly outlined below, the differences to the hitherto de-
sign approaches of 1999 being pointed out.
Fundamental design assumptions:

• Interpretations of navigation observations at the Dortmund-Ems-Canal, the Wesei-Datteln-
Canal, the River Weser at Stolzenau and recent studies at the Lower Havel Waterway (LHW) in
the Ketzin area have shown that the largest waves are generally not produced by fully laden
vessels, but by empty vessels and in exceptional cases by recreational boats and passenger
ships. Therefore, it is recommended in the GBB that all relevant vessels and loading depths
should be studied of equal importance. This is also valid for the geotechnical design since the
drawdown time - which decreases with increasing ship speed - significantly determines the
necessary revetment thickness.
• The Load stemming from the propulsion systems on bed and banks, as well as the bow thrust-
ers were considered . The appropriate formulae were taken from Schokking's Master of Science
Thesis (2002). With exception of very narrow canal cross-sections as those for one-way traffic
and very low underkeel clearances, these loads are generally less important for revetment de-
sign than those from the primary wave system of the ship: drawdown, return current and espe-
cially transversal stern waves.
• In order to enlarge the GBB application range, not only canal conditions were studied, but algo-
rithms were given additionally in order to be able to investigate wave heights in large cross-
sections, as they occur, for instance, in the LHW in lake-like widenings.

Hydraulic Load on revetments while sailing at typical speed
• In order to be able to investigate larger crosssections or - under canal conditions - short vessels,
e.g. tugs, pusher craft sailing alone or recreational vessels, the principle of the 'equivalent canal
cross-section' with cross-sectional area Ac.equi was applied, as proposed amongst others by
Söhngen (2002). Ac.equi considers ' the lateral restriction of the return flow field. Therefore exist-
ing approaches, e.g. the 1D canal theory of Schijf and Jansen (1953) can be applied for these
cases.
• To account for friction effects, the boundary layer close to the vessel forming from the bow to
the stern was considered by its displacement thickness. In this context, an equivalent ship
cross-section will be used in applying Schijf's theory that is enlarged by the boundary layer
thickness. Further friction effects that generally lead to a water level slope from bow to stern
-6-
Autorenfassung
2010
and a stern heavy trim in canals are considered by means of subsequent correction of the calcu-
lated drawdown values that have been calibrated according to the field tests undertaken at
WDC and extensive 30 model calculations.
• The above-mentioned field tests revealed that the critical ship speed does not very much de-
pend on the propulsion power and on the eccentricity of the ship path. This was mainly caused
by the fact that in the field tests - as also frequently happens in practice - no real stationary sail-
ing situation could be reached but that the vessel approached the bank in a non-stationary way
when sailing close to the bank. In a perspective of design values that ought to be as safe as pos-
sible, the impact of the values mentioned on the critical ship speed - which in principle reduce
the ship speed du ring stationary sailing close to the bank - was therefore neglected.
• In order to be able to consider vessels with a static trim, the distinction was made between bow
and stern wave heights; with homogenously laden ships the bow wave height is above all rele-
vant for the drawdown time and thereby for the geotechnical design, whereas the stern wave
height is relevant for the hydraulic design i.e. for the necessary stone size in loose ripraps. With
short vessels where a clear distinction between bow and stern wave height is not possible any-
more, the stern wave and the related drawdown time may be relevant too.
• The impact of static trim and friction was considered empirically by means of the field tests
undertaken in 2002 at the WDC. For shallow water situations, the respective presuppositions
are safe - as recent measurements on the River Havel have shown. Here surely a correction will
be needed in the future in order to avoid over-dimensioning in large cross-sections.
• Theoretical considerations - based on the wave energy and numerical 3-dimensional model
calculations for canal conditions - confirm the equations given in the PIANC publication for
shoaling of wave heights in the case of long vessels compared to the canal width sailing close to
the bank. For other vessels the PIANC formulae were not applied since the shallow water effect,
which becomes relevant here, is already considered by the equivalent canal profile.
• For drawdown time and drawdown speed, the influences of the distance to the bank and of the
ship speed were taken into account. The equations are based on a theoretical reflection on the
relevant length of the bow and stern waves. The formulae for the drawdown time were cali-
brated using the measurements at WDC. By this means, over-dimensioning by a hitherto
planned constant drawdown time valid only for very narrow canal cross-sections and tradition-
al vessels is avoided.
• The slope supply flow was considered analogously to the studies executed by Blaauw et al.
(1994). This flow occurs in case of breaking stern waves. In addition to the studies performed
by Blaauw, a function was created, relating the slope supply flow velocity - which can reach the
ship speed in case of high waves in narrow canal cross-sections - to a special Froude number
resulting from ship speed and wave height. This approach was calibrated by the above-
mentioned field tests at WDC and will be supported by new field measurements
-7-
Autorenfassung
2010

in the extremely narrow cross-section of the one-way traffic branch canal near Salzgitter and the a.
m. model tests at DST.
From dimension considerations, a practical breaker criterion was elaborated which depends on the
wave steepness formed by the wave height and by the distance to the bank as an index of the wave
length and on the local Froude number in the drawdown area. The breaker criterion is important
since the bank Load caused by non-broken waves is usually low and therefore negligible.
When calculating secondary waves, the fact is taken into consideration that, when short vessels sail
close to the bank, bow and stern wave heights can superpose each other at the bank (Fig. 3). Fur-
thermore, it can be considered in the GBB that, when recreational boats move close to the planing
speed, the ship-induced bow and stern wave systems interfere with each other. For quantifying
these effects, the analogy to the wave resistance of vessels according to Horn (1928) was used. And
empirically, by means of field tests, the closeness to the critical ship speed was also considered
where shoaling of secondary wave heights may occur. In addition, an upper limit was set
Fig. 4: Design variables calculated according to GBB of a fully laden GMS sailing close to the
bank in a German standard trapezoidal shaped canal profile, entered as a function of
the ship speed (umax,bow = max. velocity of the slope supply flow, vreturn,stern. = return flow
velocity at the bank in the stern area, Hbank,stern = stern wave height at the bank, Δ h =
mean drawdown, Hsec,transversal = secondary wave height of the transverse stern waves,
Hsec,intenerence = secondary wave height of the interference waves close to the bank)
-8-
Autorenfassung
2010

Fig. 5: Maximum secondary wave heights Hmax of planing boats as a function of bank distance
u at planning speed, calculated after Maynord (2005) for two recreational boats of a
similar geometry (draught/length = 0.06, beam/length = 0.3) with lengths of 5m and
10m for deep-water.
In order to estimate the wave heights of recreational vessels, which can sail generally above the
critical ship and also with planing speed vSpl, where the vessel 'climbs' its own bow wave and where
the waves are highest, an empirical relation given in Przedwojski (1995) was used. From this a ra-
tio for the maximum secondary wave height was deduced. According to this, recreational boats'
draught is an important determinant of the wave heights. Further impacts arise from the ratio of
ship length to bank distance - with the largest waves occurring when the vessel moves at a distance
to the bank corresponding to about half the ship length. In the example given in Figure 5 a recrea-
tional vessel of 1Om length, 0.6m draught and 3m beam width results in a wave height of some 1.1
m. This value can be compared e.g. with a formula provided by Maynord (2005), which was de-
duced from field tests with typical American fishing boats. Assuming that this formula is not only
valid for vS ≥ vSpl, but also for the semi planning speed range, resulting in a value of about 1.2m,
which is in good agreement to the GBB approach.
1.3 Impact of propulsion
1. The induced initial speed - necessary for calculating the propeller jet propagation - immedi-
ately behind the propeller or, in the case of free propellers, once the jet contraction has fin-
ished, was determined for different rates of advance J ≠ 0 from recent investigations of
-9-
Autorenfassung
2010
Peters (2002). Here the initial speed results as the jet velocity relative to the ship. This ini-
tial speed is depending on the ship speed relative to the surrounding water body vS and can
be slightly higher or lower than for a vessel that is resting relative to the water (vS = 0). As,
in addition, the sailing ship usually moves forward relative to the bottom, the relative jet ve-
locity at the bottom - that is relevant for the design - is generally lower than at vS = 0. This
can be seen in Figure 6, using the well-known approaches of Römisch and Fuehrer result to
calculate jet velocities near the bottom of a typical GMS for different ship speeds and geo-
metrical boundary conditions as they occur in free-flowing waters as in the River Rhine dur-
ing low water stages
Figure 6 on the next page shows the huge impact of the ship speed on the jet velocity near
the river or canal bed. At vS = 3m/s = 10.8km/h when moving ahead, the latter reduces to
about the half compared to the value at vS = 0. Also the propeller design has a huge impact. lf
the selected entire power of all the props is 800kW

Fig. 6: Near bed propeller induced jet velocities in stagnant water from a large motor vessel
with 8OOkW total engine power used, having different design features of the props
(one wheel, two-wheeler, props in nozzles or free wheel) and driving with different
ship speeds vs (left: starting at rest, middle: accelerating after manoeuvring, right: typi-
cal travel speed), calculated for different water depths/draughts (ts at) low stages of the
River Rhine (2.2m/1.6m) and at mean stage (4.0m/2.8m) and German canal depth re-
spectively with corresponding keel clearances kc accounting stern and squat.
as assumed and if this power is realised by 2 ducted propellers, the jet velocity at vs = 3 m/s
could be reduced from about 1.9m/s for one prop to about 1.3m/s using two wheels. With a
- 10 -
Autorenfassung
2010
free propeller, however, the jet velocity would dramatically increase compared to a corre-
sponding ducted propeller.
If for this example (propeller diameter D = 1.6 m, average keel clearance kc = 0.6m) the nec-
essary stone size D50 (grain size at 50% of particles passing) is calculated according to GBB,
using stones with a grain density of ρs = 2650 kg/m3 - on the one hand using the ratio given
by Römisch and Fuehrer (R/F, 1985) which should help to avoid scouring of bottom protec-
tions completely if possible - and alternatively using a ratio given by Ducker and Miller
(D/M, 1996) for the scour depth resulting from a short-time Load that is used conversely in
order to obtain a design formula for the necessary D50, presuming an admitted scour depth
chosen at 1 x D50 - then the necessary stone sizes can be calculated as follows:
• Ducted props:
vs =0.0m/s (ship at rest), D50 = 0.80m according to R/F, 0.18m according to D/M respectively
vs =3.0m/s (moving ahead), 0 50 = 0.27m R/F, 0.06m D/M
• Free wheel:
vs = 0.0m/s, 050 = 1.45m R/F; 0.33m D/M
vs = 3.0 m/s, 0 50 = 0.35m R/F; 0.08m D/M
This example shows huge differences in the necessary stone size depending on the design
approach chosen. Hence, a sufficient revetment stability can already be obtained using rela-
tively small stones - if it can be presumed that the propeller jet does not always have an im-
pact on the bottarn at the same point so that revetment damages cannot cumulate, and that
at the same time the revetment has been dimensioned sufficiently thick to ensure that a sin-
gle scour is not deeper than the revetment thickness.
It should be mentioned that due to the upwards-directed convergent flow at the stern - the
propeller jet can be diverged from the bottom while moving ahead. The extent of diversion
mainly depends on the form of the underwater hull at the stern. This effect which is reduc-
ing the propeller impact on channel bed should be considered in further studies to improve
the design methods in later releases of GBBSoft. Therefore the algorithms currently given in
the GBB and GBBSoft are generally on the safe side for a ship moving ahead.
- 11 -
Autorenfassung
2010

• As already mentioned, formulae were provided in the GBB for bow thruster's jet propagation.
For this, a safe approach was used regarding the jet propagation angle and thereby for the jet
velocity reduction from the vessel to the bank. For large bank distances, corrections are re-
quired in order to avoid over-dimensioning. Only few hints are given on this in the existing lit-
erature, so own studies ought to be pushed.
2 Geotechnical design- determining the revetment thickness
2.1 General aspects
According to GBB, the necessary revetment thickness dD results from the following three require-
ments that all need to be met:
• Together with the subsoil, the revetment must guarantee a stable bank, i.e. it must provide a
minimum thickness which is necessary for static reasons. It is determined by geotechnical de-
sign.
• Riprap needs to provide a sufficient stability of the layer. Thus a minimum thickness of dD is
essential. The minimum thickness is related to the size of the stone, decisive is the maximum
stone length L.
• Minimum thicknesses, for example, for protection against impacts by shipping and anchor cast,
need to be met.
The geotechnical design (criterion 1) has been fixed for the first time as a standard for revetment
design in the GBB. A crucial new aspect of geotechnical design is the consideration of the excess
pore water pressures which can be caused by rapid drawdown. The evidences required by this are
briefly outlined below and the crucial input parameters are developed. More detailed information
can be found, for example, at Holfelder and Kayser (2005).
The criteria 2 and 3 are laid down in the codes and guidelines, e.g. in the MAR.
2.2 Excess pore water pressure caused by rapid drawdown
The pores in the soil below the groundwater level are usually considered to be completely water-
saturated. lf - as in the case of a permeable revetment - there is an immediate hydraulic contact
between the body of water and the pore water, a water level change in the body of water (e.g. a
wave) would provoke an immediate pore water pressure change as the completely saturated pore
water is incompressible - see Köhler (1989), amongst others.
- 12 -
Autorenfassung
2010
Fig. 7: Flow lines and equipotential lines in the ground below a permeable slope revetment
during rapid drawdown of the water level

However, the natural pore water contains gas bubbles of microscopic sizes that may consist of air
or gases resulting from bacterial decomposition. This causes a crucial change in the water's physi-
cal behaviour. The pore water cannot be considered as an ideal, incompressible fluid anymore as it
becomes compressible by its gas fraction. This causes the gas contained in the water to expand
when there is a pressure reduction e.g. caused by drawdown during a ship passage. Therefore, a
pressure compensation in the pore water is connected with a flow process in the pore space. The
flow process delays the pressure reduction in the pore water compared to the free water level in
the body of waters, and excess pore water pressures arise. The excess pore water pressure pro-
vokes seepage flow towards the ground surface, thus reducing the local stability of the bank and/or
bed (see Figure 7).
2.3 Design approaches for revetment thickness according to GBB
In order to guarantee local stability of the waterside slope, the revetment must provide the statical-
ly necessary thickness dD. This thickness is to be determined by means of geotechnical design using
local, i.e. surface-near failure mechanisms.
For detection of the local stability, safety factors for impacts and resistances are set at y = 1.0, which
means that no safety margin has been calculated for. This is justified by the fact that local revetment
failure can be detected in good time and that therefore there would still be enough time for mainte-
nance measures. Local failure is not expected to cause major catastrophes.
- 13 -
Autorenfassung
2010
For determining local stability there are two basic failure mechanisms to be distinguished: on the
one hand, it must be shown that no hydrodynamic soil displacement is possible which may cause
longer-term damage of the revetment stability. On the other hand, the revetment dimensions must
be sufficient to provide protection against sliding along a failure surface in the ground parallel to
the slope below the revetment.
The evidence against hydrodynamic soil displacement corresponds to the proof of hydraulic failure
usual in geotechnical engineering where the holding weights G are opposed to the driving flow
forces from the excess pore water pressure ΔU, cf. Figure 8.
Fig. 8: Equilibrium of forces for the determination of hydrodynamic soil displacement
Fig. 9: Equilibrium of forces for the determination against sliding of the slope
- 14 -
Autorenfassung
2010

The evidence against revetment sliding is based on an equilibrium of forces in the boundary state.
The relevant sliding wedge is determined by the critical depth dcrit at which the effective stresses in
the soil (difference between excess pore water pressure and effective weight of the soil) become
minima. To be set are the effective weight G', the excess pore water pressure Δu, the frictional force
Q, the cohesion force C' if present, as well as the holding forces F from a toe support (Figure 9).
The parameters relevant for geotechnical revetment design are the soil permeability, the shear
strength of the soil (internal angle of friction, cohesion) and the measure of rapid drawdown. Ac-
cordingly, in the course of hydraulic design, special attention needs to be assigned to their determi-
nation.
References
BLAAUW, H. G.; VAN DER KNAAP, F. C. M.; OE GROOT, M. T.; PILARCZYK, K. W., "Design of bank pro-
tection of inland navigation fairways", International conference on flexible armoured re-
vetments incorporating geotextiles, Pub. No. 320, London, 1984.
BUNDESANSTALT FÜR WASSERBAU (BAW), KARLSRUHE 'GBB: Principles for the Design of Bank
and Bottom Protection for Inland Waterways', Mitteilungsblatt Nr. 88, 2005.
FUEHRER, M.; RÖMISCH, K., "Dimensionierung von Sohlen- und Böschungsbefestigungen an Schiff-
fahrtskanälen", Mitteilungen der Forschungsanstalt für Schifffahrt, Wasser- und Grundbau
(FAS), Schriftenreihe Wasser- und Grundbau, Nr. 47, Berlin, 1985.
HOLFELDER, T.; KAYSER, J., "Berücksichtigung von Porenwasserüberdrücken bei der Bemessung
von Deckwerken an Wasserstraßen", Proceedings Ohde-Kolloquium, TU Dresden, 2005.
HORN, F., "Theorie des Schiffes", in "Mechanik der Flüssigkeiten nebst technischen Anwendungsge-
bieten", Verlag Johann Ambrosius Barth, Leipzig, 1928.
KÖHLER, H.-J., 'Messung von Porenwasserüberdrücken im Untergrund', Mitteilungsblatt der Bun-

desanstalt für Wasserbau Nr. 66, S. 155 - 174, Karlsruhe, 1989.
BUNDESANSTALT FÜR WASSERBAU (BAW), MAR: Code of Practice 'Use of Standard Construction
Methods for Bank and Bottom Protection on Waterways', Karlsruhe, 2008.
- 15 -
Autorenfassung
2010
MAYNORD, S. T., "Wave height from planing and semi-planing small boats", River Reseach and Ap-
plications 21, 1-17,2005.
PETERS, H.-E. , "Stellungnahme und Vorschläge zur Abschätzung der Strahlgeschwindigkeit von
Propellern und Düsenpropellern", Bericht an die BAW, Rostock, 2002.
PIANC, "Guidelines for the design and construction of flexible revetments incorporating geotextiles
in marine environment", Report of Warking Group 21 of the Permanent Technical Commit-
tee II, Supplement to Bulletins Nos. 78/79, 1992.
PRZEDWOJSKI, B.; BLAZEJEWSKI, R.; PILARCZYK, W. K., "River training techniques - fundamentals,
design and application", Balkema, Rotterdam, 1995.
RÖMISCH, K., "Der Propellerstrahl als erodierendes Element bei An- und Ablegemanövern in Ha-
fenbecken", Seeswirtschaft Heft 7, Berlin, 1975.
RÖMISCH, K., "Empfehlungen zur Bemessung von Hafeneinfahrten", TU Dresden, Sektion Wasser-
wesen, Wasserbauliche Mitteilungen, Heft 1, S. 2-84, Eigenverlag, Dresden, 1989. RÖMISCH,
K., "Propellerstrahlinduzierte Erosionserscheinungen", HANSA - Schifffahrt - Schiffbau - Ha-
fen, Nr. 9, 1994.
SCHIJF, J. B.; JANSEN, P.P. , PIANC's 181h lnt. Nav. Congress. Section I, Communication 1; Rom 1953.
SCHOKKING, L. A., "Bowthruster-induced damage", Master of Science Thesis, TU Delft, 2002.
SÖHNGEN, B., 'Dimensionierung von Fahrrinnenquerschnitten im Rahmen der Planung von Stau-
stufen', 13. Duisburger Kolloquium für Schiffs- und Meerestechnik, Mai 1992.

Summary
Because there were significant changes in the boundary conditions influencing ship-induced loads
on bed and bank protections of German waterways, new field investigations and model tests were
carried out to examine and adapt existing revetment design methods. This led to a state of the art
revue of design approaches, compiled in the 'Principles for the Design of Bank and Bottom Protec-
tion for Inland Waterways' (GBB, 2004) and in a revised version of German Code of Practice 'Use of
- 16 -
Autorenfassung
2010
Standard Construction Methods for Bank and Bottom Protection on Waterways' (MAR, 2009). Addi-
tionally, the design principles of GBB, including all the experience gathered since 2004, were put
into a user-friendly software tool, named GBBSoft, which is available for engineers, researchers and
training purposes on the BAW website (www. baw.de) since the beginning of 2009, together with a
new version of the GBB, which contains all the modified approaches, used in 'GBBSoft'. The present
publication gives a brief insight in the new design approaches and application examples.
Résumé
A cause de l'évolution significative des actions génerées par les bateaux sur les berges et le fond, de
nouvelles campagnes de mesure in situ et de tests en laboratoire ont été conduites pour réévaluer
et adapter les nègles existantes de dimensionnement des protections. Cette démarche à conduit à
un état de l'art des règles de conception, publié dans les «Principes de dimensionnement des pro-
tections de berges et de fond pour les voies navigables» (GBB 2004) et a une révision des recom-
mandations allemandes « Utilisation des règles de l'art standard pour la protection des berges et du
fond des voies navigables» (MAR 2009). De plus, les principes de dimensionnement de GBB, y com-
pris le retour d'expérience depuis 2004, ont été intégrées dans un logiciel convi-vial , GBBsoft, dis-
ponible pour les ingénieurs, les chercheurs et l'enseignement sur le site internet du BAW (www.
baw.de) depuis le début 2009, avec une nouvelle version de GBB, décrivant tous les amendements
mis en ceuvre dans «GBBsoft». L'article présente succinctement les nouvelles méthodes de dimen-
sionnement et des exemples d'application.
Zusammenfassung
Wegen signifikant veränderter Randbedingungen bezüglich schiffsinduzierter Belastungen auf

Sohl- und Uferbefestigungen an deutschen Binnenwasserstraßen wurden neue Natur- und Modell-
untersuchungen durchgeführt, um bestehende Bemessungsgrundlagen für Deckwerke zu überprü-
fen und anzupassen. Dies führte zu einer Zusammenstellung des bestehenden Wissens zur
Deckwerkbemessung im BAW - Mitteilungsblatt Nr. 87 mit dem Titel "Grundlagen zur Bemessung
von Böschungs- und Sohlensicherungen an Binnenwasserstraßen" (GBB, 2004) und zu einer über-
arbeiteten Version des "Merkblattes zur Anwendung von Regelbauweisen für Böschungs- und Soh-
lensicherungen an Wasserstraßen" (MAR, 2009). Zusätzlich wurden die Bemessungsgrundlagen in
Form einer anwenderfreundlichen Software mit dem Namen GBBSoft umgesetzt, die der interes-
sierten Fachwelt seit Anfang 2009 auf der BAW Webseite (www.baw.de) zur Verfügung steht, zu-
sammen mit einer neuen Fassung des GBB, die alle in "GBBSoft" modifizierten Algorithmen
dokumentiert. Die vorliegende Veröffentlichung gibt einen kurzen Überblick über die neuen Be-
messungsgrundlagen mit Anwendungen.
- 17 -

Pub 2010

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Pub 2010

Hochgeladen von

Copyright:

Verfügbare Formate

Article, Accepted Version

Söhngen, Bernhard; Kayser, Jan

Verfügbar unter/Available at: https://hdl.handle.net/20.500.11970/100652

Vorgeschlagene Zitierweise/Suggested citation:

Powered by TCPDF (www.tcpdf.org)

Erstveröffentlichung in PIANC Magazine no° 140 (2010), S. 5-16.

Design of bank and bottom protection

By Bernhard Söhngena, Jan Kayserb

Mots-clefs: Defenses de berge, dimensionnement hydraulique, dimensionnement geotechnique, enro-

1 Hydraulic design- determining of riprap stone size

Söhngen, Kayser: Design of bank and bottom protection. S. 6

Söhngen, Kayser: Design of bank and bottom protection. S. 7

Söhngen, Kayser: Design of bank and bottom protection. S. 8

1.2 The new design principles

Fundamental design assumptions:

Söhngen, Kayser: Design of bank and bottom protection. S. 9

Hydraulic Load on revetments while sailing at typical speed

Söhngen, Kayser: Design of bank and bottom protection. S. 10

Söhngen, Kayser: Design of bank and bottom protection. S. 11

1.3 Impact of propulsion

Söhngen, Kayser: Design of bank and bottom protection. S. 12

Söhngen, Kayser: Design of bank and bottom protection. S. 13

2 Geotechnical design- determining the revetment thickness

2.1 General aspects

2.2 Excess pore water pressure caused by rapid drawdown

Söhngen, Kayser: Design of bank and bottom protection. S. 14

2.3 Design approaches for revetment thickness according to GBB

Fig. 8: Equilibrium of forces for the determination of hydrodynamic soil displacement

Söhngen, Kayser: Design of bank and bottom protection. S. 15

KÖHLER, H.-J., 'Messung von Porenwasserüberdrücken im Untergrund', Mitteilungsblatt der Bun-

SCHOKKING, L. A., "Bowthruster-induced damage", Master of Science Thesis, TU Delft, 2002.

Söhngen, Kayser: Design of bank and bottom protection. S. 16

Wegen signifikant veränderter Randbedingungen bezüglich schiffsinduzierter Belastungen auf

Das könnte Ihnen auch gefallen